FRC 694 StuyPulse | 2025 Build Thread | Open Alliance

Pev_Vail · December 27, 2024, 11:38pm

¡Bienvenidos! distinguished perpendicular people! Our members identify under the affiliation StuyPulse, FRC Team 694, from Stuyvesant Secondary School in the City of New York, and we are excited to start our FIRST OpenAlliance Build Thread for the 202*10 + (2+3) REEFSCAPE Season!!

Here’s what we’ve been up to:

Offseason Competitions

BattleCry @ WPI 24 - June 1 & 2

We always love competing at BattleCry, and this year was no exception. We entered our season robot Izzi and ran driver tryouts throughout qualification matches.

The only major robot failure we experienced was these: internal compliant wheels that passed notes between our intake and our shooter shredded when we acquired two (three?) notes at once and tried to keep them moving through without outtaking. Our handoff immediately jammed and we changed these wheels out (by taking our whole shooter off first) before the next match.

Pit note: I started using our existing Slack channel where we log changes & repairs during competition to also schedule tasks needed to be done soon but not now. We liked this!

Also at one point I covered up the time on our superpit TV. We did not like this.

We seeded 14th and entered the playoffs as the captain of Alliance 10. With our alliance partners 7407, 4575, and 8724, we made it through three rounds of playoff matches. We also had a blast eating dinner & playing lawn games with our friends on 6328 on Saturday night!

Duel on the Delaware - June 22

There was no cell service at Duel. This took out all of our slack-based robot documentation, which mostly presented itself as an issue once we were back home, trying to fix Izzi’s issues, and we couldn’t remember what happened. At future service-less events, we might start writing everything down offline and then dropping it in the competition Slack channel after the fact.

We seeded 8th and picked 369 and 9998 (2539’s second robot) for our alliance. We played a really awesome playoffs run, moving straight through the upper bracket to finals! …and then had match-losing swerve module issues as soon as we got there. Inspection in the lab later showed that one of our CANcoders’ power was cut, and replacing that wire brought back our fourth wheel. Here it is with no indicator lights, almost like it lost power or something

A favorite fix of ours for when one swerve module fails (something that happened to this robot a lot) is to swap its wheel for a free-spinning omni wheel. If it’s not gonna run, at least it won’t drag.

OFFSEASON ROBOT!

With the rest of our summer, we built a clone of 2056’s CRESCENDO robot, LOW KEY, which we named Power of Friendship. We spent a lot of time working in the NYC FIRST STEM Center to assemble and test our robot. You can read about that robot in our post to 2056’s tech binder thread here.

Chezy Champs - September 27, 28, & 29

This was the first competition for our offseason robot! PoF came to Chezy with about 2 autonomous routines that both almost worked. We re-learned a lot about the value of testing time. Although we didn’t perform to the best of our ability, this competition really reminded us to keep pushing forward and served as a motivation to try harder for our 2025 Reefscape season!

…We will remember to tighten nuts.

NYC Robo Replay - October 19 & 20

This event’s best feature was its short competitor list, giving each attending robot ~1 million qualification matches. Every member of our team got tons of valuable practice, in the pit, strategy, and drive team. We ran our newly built offseason robot and our onseason robot to give twice as many students the opportunity to gain all that experience!

Power of Friendship’s (PoF or pof) pit crew kept using the Slack channel task scheduling and it was really nice. Time was so tight in our match turnarounds and we felt like we really had a handle on things. We also started emoji-reacting tasks’ status instead of threading with a “resolved” or “done.” It’s just faster.

We got to play one qualification match with both of our robots as well as our friends on 1796 the RoboTigers in the same alliance. We had a lot of fun and managed an event high score of 129 points!

This was our competition robot, Izzi’s, last event, and it really died how it lived: throwing buckets of swerve module errors all through quals and then immediately locking in for playoffs to become the lower bracket grim reaper.

Thanks to both robots’ alliance partners from 1796, 8739 (AND 8739’s second robot!), and 369 for a really fun competition.

Brunswick Eruption - October 26

After an insane FRC season, we ended it off with a bang as Power of Friendship went undefeated the entire competition 11-0.

PoF struggled to drive in one playoff match. When any one CANcoder’s power was tugged, all four Mk4i CANcoders were affected. We found that the row of WAGO terminals on our REV MPM were loose, affecting the CANcoders’ power when they were jostled. We hot glued the connectors into the MPM housing and had no further issues.

Thanks to our alliance partners from 1796 and 369 for one last successful and fun competition in 2024.

Newbie Education

We had over 200 sign ups for our team this year with 120 signups for Mechanical Engineering, 50 for Software Engineering, and 30 for Business.

For mechanical engineering, during the summer, we focused on creating a standardized newbie education curriculum consisting of newbie games (where newbies design, CAD, and prototype a specific mechanism for a game piece from a previous year), tools education, basic CAD education, and sub department education. During the school year, our newbies collaborated and created their own mechanism while learning more about the engineering design process. Here’s our curriculum:

For software engineering, we taught the basics of programming in Java (operators, variables, functions, etc.), how to use Git and GitHub for version control, and moved on to OOP concepts (inheritance, abstraction, encapsulation, etc.). We ended off by teaching the fundamentals of robot programming and how to use WPILib (programming commands, autons, subsystems, button binding, deploying and running robot code). We reinforced these concepts with project-based learning (i.e. our MotorLabs, where newbies program basic logic for controlling motors, and our ShooterLabs, where newbies learn how to program a shooter, one of the most basic subsystems in FRC). Our curriculum can be found on our GitHub:

Goal Setting

We held 2 goal setting meetings these past couple of weeks. For the first goal setting meeting, we put random ideas on a gigantic whiteboard, discussed, and sorted through each of them, giving us one final list of all the goals we wanted to achieve during the season and off season.

Our second goal setting meeting was oriented around setting tasks for us to complete to get set up for the new season. From the second goal setting meeting, we created a final to do list for things to keep in mind as we approach 2025 REEFSCAPE.

Mock Kickoff

The general Kickoff format for us is we all read the game manual together. We then split off into small groups where we analyze and discuss the functions, rules, and strategies of the game, and then regroup at the end to do one final run through of the entire game. We generally avoid talking about mechanisms during the first day of Kickoff.

This year’s mock kickoff used the 2019 game. We gave each small group a specific topic to focus on, so that when we all regrouped at the end, there would be a group of people ready to provide information on their respective subject areas. These include:

Score breakdown
Auton
Possible Alliance
Possible robot archetypes
Defense
List of Tasks
Priority list
Anything else specific to that year’s game

From there, we reviewed how accurate our predictions were in reference with how the game played out that year.

StuySplash

On Saturday, December 14th, we held our annual lecture series, StuySplash, with over 300 attendees. Lecturers from teams all over New York came to share their expertise on engineering, strategy, and team welfare.

Lecture videos will be available on our YouTube channel and our website soon. Past talks are already up - there’s lots to learn from them if you’re interested!

Anyway thanks so much for reading everyone we’re SOOOOO EXCITED for 2025!!!

Written by: Esther, Ethan, Jaco, Rui, Kalimul, Ian, Mustafa, Mr. Blay, and Pev

jxia · December 27, 2024, 11:38pm

elin-semicolon-D · December 27, 2024, 11:42pm

soriiku17 · December 27, 2024, 11:45pm

soriiku17 · January 6, 2025, 3:56am

KICKOFF

The first rule of kickoff day 1 is don’t talk about kickoff day 1. The second rule of kickoff day 1 is NO MECHANISMS. Our goal for the day was not to dive right into designing mechanisms and planning out autons but to thoroughly understand the game.

After analyzing the game manual, we broke off into groups for small group discussions on different aspects of REEFSCAPE. Our focus for these discussions was to break the game down. We highlighted points such as scoring breaking, list of tasks, robot and alliance types, and what kind of robots would win regionals and champs.

After a bit of discussion, we all came back together as a team to share what we talked about in the smaller groups. Each of our groups had prioritized analyzing different parts of the game.

We clarified some ambiguous rules, discussed possible strategies, and did our favorite kickoff activity: human match.

Human match is exactly what it sounds like. We had taped the entire field out and placed people into desk chairs to simulate a match. We tested different types of autons and defense with this and it also gave us a great idea on what teams could do in matches. We absolutely loved the complexity and infinite combinations of strategies in this game!

Towards the end of the meeting, we discovered something interesting that could affect robot strategies. A regular drivetrain can push algae into the processor (kinda)!

Additionally we did some very scientific experiments and tested how algae bounces when dropped from roughly 10 ft onto field carpet.

|2047.7016742770168x1535.6669540336236

Sunday

Day 2 is when we ordered our tasks, made a priority list, and discussed our mechanism ideas!

As of writing, we are still debating between different gear ratios for our WCP X2is and whether to prioritize robot acceleration, high top speed, or reducing tipping and are still deciding what combination would best suit our needs. We really don’t know where we fall on the gear ratio decision; we may still want to be in the L3 range.

|2048x1535.7996130255528

Choosing Mechanisms:

These are the final mechanisms we chose to prototype!

Coral Manipulation:

roller claw + wrist with degree of freedom: The mech we are prototyping is a roller claw that has a wrist allowing it to intake coral horizontally and score it either vertically or horizontally (on a pivoting arm and lift) (bottom right in following image)
barrel prototyping: We will have a box/cylindrical tube that rotates around.(top left in image)
Coral funnel claw on a virtual four bar: this is an end effector mechanism where we will intake coral horizontally and then funnel/reorientate them into a vertical position. The end effector will be on a virtual 4bar arm to control the angle of the end effector for the intaking/outaking.
A fixed chute on an elevator: This idea involves making the chute static on the elevator. The mechanism spans across the whole robot, where we intake from coral station vertically from one side and score on the other. There would be a conveyor that leads from one side to the other. The biggest benefit here is that we don’t need an additional degree of freedom (like the wrist claw), which simplified our design, and simplicity is our goal. We will be sure to test for optimal intake angle and angle

Hunting algae:

Slap down intake: An intake that “slaps down” on the algae on the ground against the bumper. The algae can be stored at the pivoting point and released against the processor, and the algae will be scored. We want to try and make this mec and the ball take up as little robot real estate as possible. We are looking at pinching it between the inner roller and the bumper and dangling it mostly outside of our robot.

Climb:

At this point we have decided that our climb would mostly be in the center of the robot, and the algae intake, along with the coral scoring mechanism, will be on opposite halves (actually thirds?) of the robot.

Fork Climb: We will cut a massive (not really) square into our belly plate to fit the cage through. By driving into the cage, our v shaped roller lines will catch the cage and force it into a single, consistent position. At this point, 2 “Forks” will come down into our drivetrain and push on the bottom plate of the cage, which lifts the robot up. This method really gives electronics a pain, but seems functional and plausible. Worth a try!

Why not tilt the entire robot? (Fork Climb): This project involves a fork-hook type clip with a claw on top to intake the cage and have the forks stick into the cage. When we climb, the fork-claws press down and hook on the edge, allowing us to flip our drivetrain up. Inspiration from: 1678 in 2023 and 1114 in 2010 (1678:https://youtu.be/9AZHA0hNYrY?si=jMjAIc1HswvSHm3L , 1114: https://youtu.be/N-Nf85hDx4Y?si=noR_5FWGMPPGfg3L )
LEGOOOO MOVIE (Lego Climb): We will be Prototyping a climber for the deep climb based off of the Diagonal lift mechanism for Deep Cage. The robot will run into the cage and set it to lie at an angle so that the 2 pinchers on the side can grab the cage bars. The robot will finally be able to lift itself up with a stem/shaft that pushes down at the bottom of the cage. We liked how the robot didn’t need to rotate that much relative to the ground, and we have high hopes for this one! (link to video: https://www.youtube.com/watch?v=Zacab9N_f-M)

Winch arm climb :We’re going to be prototyping some sort of climber arm that uses a winch to pull the cage towards the center of the robot, hopefully raising us off the ground. Inspired by: https://youtube.com/shorts/V20v1Vyutsg?si=uJh3Uaat-dLTwOsn. We will test Rollers, passive aligners, magnets, and more to find which one is the best and most consistent at grabbing the cage.

We’re so excited for this season and can’t wait to share our progress!!

Thank you to Ethan, Philip, Pev, Brendon, and Coach Blay for helping write!!
Thank you to the entire team for making this amazing kickoff weekend happen!!
And of course, thank you, reader, for following our progress!! <3

ditty · January 6, 2025, 4:04am

Very interesting stuff you got going on; I’m excited to see your guys’ prototypes!

AronK · January 6, 2025, 4:21am

Interesting designs and strategies you guys came up with. Can’t wait to see what you end up building and what your strategy will be!

jblay · January 14, 2025, 3:57pm

Week 1: Prototyping

The first week of build season is always a huge rush of activity. Everyone is excited with the release of the new game, and eager to test out new ideas and get to building a new robot. For StuyPulse, this week was a massive push to test as many ideas as possible and acquire as much data on each one. All of this is to aid us in deciding the best possible mechanisms to put on our robot.

Before diving into prototyping, we finalized our priority list for the season. This priority list was based on the tasks previously sorted into “Must Do”, “Want to Do” and “Don’t Do” (see previous post for more details).

General Ranking:

Move
Intake Coral
Store Coral
Score Coral L4
Score Coral L1
Score Coral L2
Score Coral L3
Alignment (automation) / score (automation)
Deep Climb
Knock-off Algae (Level 2)
Knock-off Algae (Level 3
Pick up and score on opposite sides
Intake Algae (Ground) (Start of want to do list)
Score Algae (processor)
(new!) Intake coral (ground)

Tasks 1-12 are ones we decided are “Must Do”; these are the tasks that we believe have to be on our robot no matter what. Tasks 13-15 are from the “Want to do” list; these are things we would like to have on the robot, but are a lower priority, if push comes to shove, we don’t have to do them to meet our goals.

We decided as a team to put scoring coral on L4 as the highest scoring priority for a few reasons. First, we knew we wanted to score on all four levels; we figured that having the ability to score L4 probably means we’d be able to score L1, L2 and L3 with some tweaks. Second, L4 gives you the most points by a considerable margin, especially in auton. Third, we ideally want to score as many L4 pieces in auton as possible and really design our robot around that ability; having that aspect of the robot be fast and stable is critical.

The alignment and scoring automation is something we’ve been including in our priority lists since 2019. It’s more so to serve as a reminder how important it is to figure out correct camera and sensor placements, as well as an indicator on when to get the robot to software.

There are some discrepancies between our previous sorting, and our final priority list. The first thing is moving shallow climb from “Must Do” to “Don’t Do”. During our priority list meeting, we talked through the possible situations where we’d do the shallow climb over the deep climb. The conclusion was that if we ever decide to not go for the deep climb in a match, we would rather spend the time scoring until the last possible second and then parking. With the deep climb being a Must Do, we decided that the shallow climb did not make sense for us to prioritize at all (we also didn’t come up with any prototypes for it, which made this decision a little bit easier). In addition from seeing some ri3d videos and some of our own testing, it seems like the deep climb is a simpler challenge than we originally assumed.
The second difference was moving Intake Coral (Ground) from “Don’t Do” to “Want to Do”. Through our initial prototyping, we saw some potential changes we could make to our algae mechanism that could possibly allow us to pick up algae from the ground. After some discussion, we decided that having that ability to clean up fallen game pieces would be nice, but is not something we want to invest a ton of time and resources into, hence its place at 15th on our priority list. It is unlikely that we end up with this ability on our final robot, and if we do it will likely only be for L1 scoring, an add on to our algae mec.

Along with our priority list, new this year for our team was a list of Import Robot Features. These are things that we wanted to keep in mind for our robot as a whole, especially during the design process. In some instances, it may also impact our final mechanism choices. The list is as follows:

Important Robot Features:

Durability
Pickup and score on opposite sides
Coral scorer as light as possible
Light lift (1 stage if possible)
Accessible electronics
High acceleration, low top speed
Form COG around Climb
Low COG
Low bumpers
Heavy
Good cameras
Passive alignment for coral?

Now, jumping into prototyping. By the time of this post, each prototype has gone through several iterations. 694’s goal with prototyping is always to gather as much information as possible, and determine the limits of each mechanism. By the end, we start comparing each prototype to each other, and narrowing down our choices. We aim to solidify our coral mechanism ASAP in order to get an alpha bot up and running by the end of week 2. I will be posting updates on each of our grouping of mechanisms as our videos for the mechanisms upload.
Written by Navid Kashem

jblay · January 14, 2025, 3:57pm

Algae Mechanism:

Going into prototyping, we had several priorities for our algae mechanism:

Take up as little depth into the robot as possible in order to free up space for the deep climb.
Be able to quickly pick up algae from the ground.
Score quickly into the algae.
Store algae without bringing it into the robot.
Dislodge algae from L2 and L3.
If possible, feed the algae mec directly from L2 and L3.

Our algae mechanism prototype has so far been really promising. We’ve found that 2-3 inches of compression works best for us, and that to securely hold the algae, we need to bring it in about 5-6.5 inches.

We also tested angles and distances from where we can score into the processor. As of right now, the mechanism will take up about 7in of depth when attached on the frame of the robot. The scoring is not exactly good enough just yet, when we are pressed up against the processor the coral gets stuck, it does score if we back out, but we are worried that that will open us up to some more misses.

The next steps for this prototype is to test some more with a bumper (potentially on a test chassis) to simulate robot scoring more accurately. Afterwards, we plan to see if there’s anything we can change to make it also be able to pick up coral from the ground and score it L1, right now it does not seem possible.

We also ran some tests with dislodging the algae directly into the mechanism from L2 and L3 and are very pleased with the results. We expect to see this ability on the final bot.